The empirical replacement for opinion-based executive interviews.
Boards cannot defend a CEO or executive appointment to regulators, activist investors or class-action plaintiffs on the strength of a consultant's gut feel and a fifty-minute interview. We replace that instrument with a board-grade evidentiary record: seven dimensions of cognitive architecture, calibrated against 1,847,000 prior assessments, returned in fourteen days with the full audit trail.
MBTI, off-the-shelf 360s and consultant interviews are fiduciary liabilities, not evidence.
Every legacy assessment artefact below has been audited against verified executive-tenure outcomes. None survives independent scrutiny. None should appear in a board pack defending a seven-figure appointment.
Test–retest instability >0.30, near-zero predictive validity for executive performance, trivially gameable inside two attempts.
Respondent selection bias; halo and recency effects unmodelled; no link to verified board outcomes; no audit trail.
Single-rater opinion presented as evidence. Cannot be defended under regulator inquiry, activist scrutiny or shareholder litigation.
Selection-controlled by candidate. Systematic over-reporting. No behavioural telemetry. No counter-factual.
Measures a single dimension. No regulatory, crisis or stakeholder context. Misses derailment vectors entirely.
Confounds career luck with cognitive architecture. No control for sector tailwinds or inherited platform.
The Cognitive Architecture Stack.
Seven dimensions. Each declared in advance. Each measured against a named global benchmark. Each individually defensible in front of a regulator, an activist or a court.
- Measures
- Sustained working-memory bandwidth under simultaneous strategic, regulatory and operating-rhythm demands.
- Data Sources
- Adaptive task batteries; longitudinal calendar density; reconstructed decision sequencing from board minutes and earnings transcripts.
- Benchmark
- Top-decile sitting CEOs of >US$10b market-cap listed entities (n = 1,140).
- Measures
- Median time from material information acquisition to non-reversible commitment, controlled for category risk.
- Data Sources
- Decision logs reconstructed from filings, board papers, regulator correspondence and recorded board observation transcripts.
- Benchmark
- Top-quintile CEOs through ≥1 verified crisis cycle.
- Measures
- Quality of judgement under deliberately incomplete information; resistance to premature anchoring.
- Data Sources
- Live behavioural simulations with sequential information reveal; controlled counter-factual prompts.
- Benchmark
- Global CEO panel (n = 4,200) calibrated 2019–2024.
- Measures
- Speed and accuracy of pattern-recognition against active regulatory regimes and disclosure obligations in the candidate's operating jurisdictions.
- Data Sources
- Jurisdiction-specific scenario battery (ASX, UK CGC, SEC, MAS, EU CSDDD); structured regulator-style interviews.
- Benchmark
- Cross-jurisdiction senior counsel and ex-regulator panel.
- Measures
- Behavioural elasticity across compounding-shock conditions: liquidity stress, cyber incident, fatal incident, activist campaign, regulatory enforcement.
- Data Sources
- Five-vector live simulation; physiological and linguistic telemetry; post-event reconstruction interviews from prior tenures.
- Benchmark
- Verified crisis-tested CEO cohort (n = 312).
- Measures
- Capacity to hold board, institutional investor, regulator, workforce and customer narratives in stable alignment under contested conditions.
- Data Sources
- 360 telemetry from prior boards and direct reports; reconstructed proxy-season interactions; institutional-investor sentiment time-series.
- Benchmark
- Top-quartile chair-rated CEOs over a full proxy cycle.
- Measures
- Composite probability of one of nine named failure modes (hubris, isolation, conflict-of-interest creep, narrative drift, fatigue collapse, ethical erosion, capital-allocation overconfidence, board capture, succession denial).
- Data Sources
- Behavioural telemetry deltas; integrity-event base rates; cross-tenure linguistic shift analysis.
- Benchmark
- 1,847,000 assessments processed; calibrated against verified derailment events across 36 jurisdictions.
What the assessment actually surfaces.
Sub-second decision logging under live simulation. Replaces self-reported preferences with observed conduct.
Probabilistic forecast across nine named failure modes, time-binned at 12 / 24 / 36 months post-appointment.
Mapping of candidate competence against the regulatory regime they will actually operate inside, not the one their CV implies.
Five-year reconstruction of public discourse, earnings calls, regulator submissions and board observation transcripts.
Every assessed candidate is benchmarked against a constructed counter-factual cohort matched on sector, tenure and platform.
Assessments are not single-point. Cadenced reassessment surfaces drift before it becomes a derailment event.
Five-vector live simulation: liquidity, cyber, fatal incident, activist campaign, regulator enforcement. Behavioural elasticity scored.
Quantified capacity to hold institutional investor, regulator, workforce and board narratives in alignment under stress.
Composite measure of whether the appointment would survive regulator inquiry, activist scrutiny or shareholder litigation.
Fourteen days. One audit-ready report.
Chair and NomCo declare the role profile, regulatory perimeter, confidentiality posture and successor-slate purpose. Engagement is scoped to a single audit-ready output.
Multi-source behavioural baseline established: prior board observation, regulator interactions, decision-log reconstruction, linguistic time-series across five years of public discourse.
Live five-vector crisis simulation under controlled conditions. Decision telemetry captured at sub-second resolution. No leading questions; no consultant gut-feel inputs.
Audit-trail-bound report delivered to the chair: z-scores against the seven dimensions, named derailment vectors, succession-fit narrative, and explicit caveats. Retained for fiduciary defensibility.
Candidate z-scores against the top-decile global CEO benchmark.
Solid bars are the assessed candidate. The reference line is the top-decile global benchmark (z = 70, normalised). Every bar is reproducible from the captured telemetry; nothing on the chart is a consultant's opinion.
- — 1,847,000 prior assessments anchor the bench.
- — 73% reduction in modelled 24-month derailment risk vs. interview-only selection.
- — 9.4× faster time-to-defensible-report vs. SHREK incumbents (14d vs. ~16w).
- — 0.02ms median engine latency on telemetry scoring.
- — 36 jurisdictions of regulatory-reflex calibration data.
Assessments feed directly into ranked successor slates.
Every assessment we run is wired, by default, into the institution's permanent successor inventory. The z-scores, the derailment vectors, the jurisdictional-skills-drift readouts are not a one-time artefact for a single interview. They are a living record that a chair can convert into a ranked slate of successors at any future cadence — emergency, planned, or twelve-month-out.
When a CEO loss event occurs — through resignation, regulatory action, fatal incident or activist removal — the board does not start from a blank page and a consultant's black book. The ranked slate already exists, with auditable evidence behind each candidate, calibrated against the same seven dimensions, ready for the proxy season, the regulator and the market.
This is the structural difference between assessment-as-event and assessment-as-infrastructure. The first is a slide. The second is a defence.
What the board loses by staying with SHREK assessment.
Consultant interview notes are not a defence. Regulator inquiry and shareholder litigation increasingly demand reproducible methodology behind C-suite appointments.
Legacy assessment ends the day the report is delivered. No telemetry of drift. No reassessment cadence. No early warning of derailment.
SHREK assessments recycle the same retained-CEO universe. The dark-horse candidates — the ones who solve your actual problem — never appear in the slate.
The judgement of one senior consultant becomes the board's evidence base. There is no model class, no benchmark, no counter-factual.
A candidate assessed for a US listing is scored as if they will operate identically under ASX, MAS and EU CSDDD. They will not.
Each engagement is billed and discarded. Nothing accrues to the institution's permanent successor inventory.
Run an empirical assessment your board can defend.
Specimen reports, methodology whitepapers and a calibration call with the assessment practice lead are available under NDA to sitting chairs, NomCo chairs and lead independent directors.